It is often assumed that cash flow affects dividend payout. This study provides evidence on the incremental information content of cash flow numbers over Profits and Previous Year's Dividends (Lintner's model) in explaining changes in cash dividends. It further examines whether different measures of cash flow differ in information content for dividend-increasing and dividend-decreasing firms. Lintner's model of dividend changes is robust across firms with either dividend increases or decreases. The null hypotheses, that no definition of cash flow adds to the model, could not be rejected for any of the definitions.